10 Ways Scientists Lie About Drugs


Does the FDA keep us safe?

Research into new medications is how we learn if they work and if they’re safe.  It’s the only protection we have against snake oil and worse.  But disasters like Thalidomide and Vioxx – not to mention the dozens of products recalled monthly – remind us that the FDA screws up every now and then.

That’s because drug companies often pretty much lie to get their drugs approved.  Here’s how.

1) Report results that didn’t happen

Amazingly enough, some drug companies take studies that show their product doesn’t work, and they report that it does!  This has happened at least 11 times with studies into antidepressants.

Another form of this is to write a conclusion that’s different from what the actual results show.  Because of time pressure, doctors don’t have time to actually read all the papers thrown at them, and often just read the abstract, which contains the pro-drug message like:  “Conclusion:  Fake drug has been shown to be safe and efficacious in treating extreme bladder discomfort.”

2) Use faulty statistics

There’s a reason Benjamin Disraeli said “there are lies, damn lies, and statistics.”  The sheer amount of ways to manipulate the data from studies by statistics is mind-boggling.  The simplest is to report that something is statistically significant when that doesn’t mean anything.

For instance, a drug can have a statistically significant effect on blood pressure – but that effect is tiny and meaningless.

3) Use inappropriate measuring systems

Some things are easy to measure, like someone’s height or weight.  Others, like mental states and attitudes, are a lot harder.  Psychological assessments range from solid to extremely shaky, like the infamous Rorschach blotch test.

Thank God no drug has been approved because of results of a Rorschach blotch test.  But drug companies always use the measuring system that puts their drug in the best light, even if it doesn’t mean it actually works.

4) Don’t use placebo

When you don’t test your treatment against placebo, absurd things happen.  Like the gastric freezing procedure to treat ulcers, which consists of putting a balloon full of frozen liquid into the stomach to cool it off.  It worked great – until it was tested against a placebo, or a fake treatment.

Guess what?  The placebo worked better.

5) Don’t test against other drugs

You’d have to be crazy to forget about testing against a placebo.  But what drug companies almost never do is test their drugs against other drugs that already exist.

Vyvanse, for instance, is a new medication for ADHD that is chemically identical to Dexedrine.  It’s a hell of a lot more expensive, so there’s no way that it’s ever going to be tested if it’s actually better to the treatment that already existed.

6) Ignore the FDA

This is by the far the stupidest thing a drug company can do.  You’d have to be crazy to ignore what the FDA tells you to do in testing your medication.  But that’s exactly what Sam Waksal did with the breakthrough cancer drug Erbitux.

They told him to conduct certain studies and change some of his protocols.  He didn’t.  The worst part is, Erbitux was a drug that could help treat cancer in some of the worst cases.  Then Sam was surprised when his drug was rejected!

7) Pressure patients to give the results you want

If you give patients a test at the start of the trial, like a mood inventory scale, and then again at the end of the trial, they won’t answer the same.  Their past experience biases how they respond the second time around.

Also, you can subtly pressure participants to give results you want by leading questions.

8)  Test it for short periods of time

Take a drug that needs to be taken for months if not years, like an anti-depressant.  Why not just test it for a few weeks?  Not only will you save money, you won’t have to learn about the nasty side effects caused by long term use.

Everyone wins, except the patients.

9) Test it on a few people

Who cares if your drug will be taken by millions?  Test it on only several hundred people.  That makes it even easier for you to run the test again if you are so unlucky as to get bad results.

10) Test it on people who’re different

Test your drug on people who are healthier than those who will eventually take it.  You’re almost guaranteed better results.

Prozac, for instance, while now used for dozens of reasons including pain management, was initially tested on mild-to-moderately depressed people – not people who were extremely depressed.  Stuff like that explains why antidepressants don’t work nearly as well as they’re supposed to.

You Might Like:

Eat Simpler, Be Healthier

50 Ways to Make Friends

Do you trust the drug companies?  Why or why not?

4 thoughts on “10 Ways Scientists Lie About Drugs

  1. Reading this was fun and enlightening to me Jane…. Why does David need to be qualified? He sheds alot of knowledge and insight on his subjects. Good job David! I enjoyed the reading very much.

  2. These debates always remind me of those 1940’s films of people invited to watch an atomic explosion. The scientific community told them it was safe! and look,they got to keep those “neat” sunglasses, besides, they are 5 miles away! Unfortunately,”ALL” those people are DEAD now, and they didn’t die of “old age”, if you get my drift. But today we are so much more careful,you know,with the genetically altered food and the M.R.I at the hospital. Maybe that’s why they call a DR.office a “PRACTICE”. Wake up folks!

  3. In response to Jane,

    Actually, many of the points the author makes in this blog post are true and can be verified as such if one bothers to reference check (e.g., antidepressant and vioxx studies), and the rest are common sense points that those of us who ‘do’ science are well aware of and are trained to avoid.

    Of course, human beings are imperfect. And as much as scientists attempt to engage in unbiased researching, if there are million dollar grant funds in the balance, it rarely ends up that way. It is even more complex when you look at the fact that drug companies are providing ALL the money to researchers in an attempt to prove that their drugs are helpful, and there are virtually no funds from unbiased sources that would support, for example, testing alternatives, or, comparing generic vs newly formulated drugs.

    I would applaud the author for making these points accessible to readers who don’t have scientific backgrounds, or those who are unaccustomed to critically appraising new information. It is such questioning that fuels more research and gets us closer to a more valid result.

    Quite frankly, scientists in general have done a terrible job in making some of the bigger issues in science accessible to people without the background. This blog I think does an outstanding job in that regard. If you are looking for something written more ‘professionally’, go read a research journal.

  4. David,

    I’m going to add another comment. I like that you have a blog about health and I’m assuming that it’s because you want people to be healthy. And that there is some connection between that and – and what? Are you a medical student? What exactly is your background that qualifies you to write about this sort of stuff?

    Let’s say you’re interested in people living healthier lives and that this is the basis of your blog. But there is a lot to your writing, and hence your blog, that is not particularly scientific. Especially when you write about how drug companies lie with and manipulate statistics.

    Let’s take a look at your post: You begin “…drug companies often pretty much lie” to get their drugs approved. Now what does “often” mean? Statistically? how often is often? And what does “pretty much lie” mean? If you’re writing an article on stem cells, you’re not going to be writing something “often pretty much” happens this way or that. So I think you need to approach your blog more professionally.

Leave a Reply

Your email address will not be published. Required fields are marked *